Fingerprint matching minutiae comparison

hello i am a student and working with fingerprint matching project..first i have extract tow minutiae from fingerprint image ..now i am compare both extraction ..but problem will arise ..if i browse different image in both axes the compare result will show image 1 and image 2 are not equal...but when i am browse same image the message will same ..what is the problem in my code ...

% --- Executes on button press in Recognation.
function Recognation_Callback(hObject, eventdata, handles)
% hObject    handle to Recognation (see GCBO)
% eventdata  reserved - to be defined in a future version of MATLAB
% handles    structure with handles and user data (see GUIDATA)
if isequal(handles.axes1,handles.axes2) % image 1 and image 2 are compared
    msgbox('Both image 1 and image 2 are equal', 'MESSAGE');
else
    msgbox('image 1 and image 2 are not equal', 'MESSAGE');
end

Answer 1

[dpb]

That will simply test two axes handles to see if they match; since you've got two different images the answer to that question is always going to be "no".

It's the content of some features within the image that you're interested in; you'll have to have isolated the features and categorized them in some fashion first; then you can do some comparison of those statistics/measures.

[Image Analyst]

Take note Indrani that this does not mean simply using isequal() to compare two images. That will never work unless the images match perfectly - every single pixel has the exact same gray level. That is never going to happen in the real world. Even if you take two pictures of a static object, say your office, and compare them, there will always be pixel differences due to normal photo noise. Because a fingerprint also has motion smear and different fields of view, the situation is even worse. That's why you need to look here: http://www.visionbib.com/bibliography/contentspeople.html#Face%20Recognition,%20Detection,%20Tracking,%20Gesture%20Recognition,%20Fingerprints,%20Biometrics and see how they construct feature vectors and compare the feature vectors instead of the images themselves directly.

[dpb]

Thanks for the cleanup and follow up...I don't do image processing so generally don't try to answer related questions but the general gist of what was wrong approach in OP's query here was well above the details of how to actually try to extract features and make the comparisons in that the handle to an image figure is only remotely connected to the content of that image.